CPT calibration chamber testing on tailings

The cone penetration test (CPT) is a widely used technique for on-site investigations. In the field, interpretation correlations of CPT can be established using laboratory calibration chamber tests. Empirical and theoretical correlations were developed for clay and sand under undrained and drained conditions, respectively. However, interpreting these results can be more challenging for intermediate soils that exhibit partial drainage during penetration and lie between fully undrained and drained conditions. Tailings, which are the by-products of mining and are often rich in fine particles, are a typical example of such intermediate soils. To gain a deeper understanding of the behavior of the tailings during CPT, a calibration chamber test was performed at a standard rate of 2 cm/s and subsequently at a slow rate of 0.02 cm/s, and the obtained results were analyzed.


Introduction
In response to the challenges associated with the sampling of cohesionless soils, such as sand, the cone penetration test (CPT) has emerged as a widely used technique in geotechnical engineering because of its continuity, efficiency, and repeatability.According to ASTM guidelines, a penetrometer tip featuring a conical point with a 60 apex angle and a cone base area of 10 or 15 cm 2 is advanced through the soil at a constant rate of 20 mm/s.During the process, the tip resistance qc, sleeve friction fs, and pore water pressure at the cone shoulder position u2 are measured.
Several empirical correlations have been established that relate state parameters to CPT measurements based on calibration chamber tests [2] [3] [5] [7] [13].However, these interpretation methods have been obtained for clays or sands, and their applicability to intermediate soils, such as tailings, remains unclear.In sand, the soil response is typically considered to be fully drained at a standard rate, whereas it is fully undrained in clay.Partial drainage is likely to occur in intermediate soils because of variations in the drainage conditions.The rate effect significantly influences partial drainage, which is primarily investigated through centrifuge testing [6] [11] [14] and field CPT [4] [10] [15], Only a limited number of calibration chamber tests have been conducted on tailings [1] [12], owing to the challenges associated with handling large-scale samples.This study introduces a calibration chamber at the University of Toronto and provides detailed procedures for preparing tailing samples, conducting tests, and providing comprehensive details.Finally, an analysis of the calibration-chamber test is presented.

Overall
A calibration chamber, previously owned by Golder Associates (WSP) and located at the University of Toronto, was recommended in 2018 [8].This chamber featured several additional components, including a reaction frame, hydraulic jack, control panel, water de-airing system, and data acquisition equipment, as illustrated in Figure 1.It has an inner diameter of 1.4 m and stands at a height of 1.14 m, with a chamber wall approximately 1 cm thick, serving as a semi-rigid boundary.Within the chamber, the samples were prepared and sealed beneath the top latex membrane, with both the top and bottom valves facilitating drainage during consolidation.

Control panel
The control panel consists of an array of pipes and valves that regulate the pressure exerted on the sample, as illustrated in Figure 2. Three air-water reservoirs, labeled AW1, AW2, and AW3, supplied water pressure to the sample.This pressure was adjusted using air pressure regulators located on the panel, accompanied by a small digital screen that monitored the pressure in each reservoir.Vertical pressure was applied to the membrane surface through AW1, whereas AW2 and AW3 applied back pressure to the bottom and top of the sample, respectively, during consolidation.The water levels in the reservoirs were measured using nearby water-level tubes on a panel.

Hydraulic jack
Owing to spatial restrictions in the laboratory, only half of the chamber extended above the surface, requiring the hydraulic jack to be lifted to the reaction frame for installation using a crane.During the calibration chamber test, a computer-controlled hydraulic jack pushed the cone into the sample at a userdefined rate ranging from 0.002 to 3 cm/s, with a maximum total force of 50 kN.enhance heat dissipation, a cooler was connected to the oil tank of the hydraulic jack, which is particularly useful when the pump must function for an extended period.A standard cone, with a diameter of 35.7 mm, was used for the test.The CPT measurements, including the tip resistance qc, sleeve friction fs, and pore water pressure at the cone shoulder position u2, were recorded during penetration.The penetration depth was measured using a pull-rope displacement sensor.

Material properties
The tailings were obtained from a gold-tailing storage facility located in Canada.The most representative gradations within the chamber, determined using a combination of sieve and hydrometer analyses, are shown in Figure 3.These tailings consisted predominantly of fine particles, with 72%-82% passing through the No. 200 sieve.The specific gravity Gs of the tailings particles was measured to be 3.2.

Sample preparation
Prior to sample preparation, the chamber was cleaned, and frictionless sheets were affixed to the chamber wall to minimize friction.A 5 cm thick layer of sand was positioned at the base of the chamber, followed by a geotextile layer and steel plate.The tailings sample preparation commenced with slurry deposition.Initially, the tailings were thoroughly mixed in a concrete mixer with additional de-aired water, and vacuum application aided the removal of air bubbles, as shown in Figure 4. Subsequently, the slurry tailings were conveyed to the chamber via a pipe, following a consistent movement pattern to ensure uniformity.The weight and water content of each layer were controlled to maintain consistency.The excess water above the sample surface was drained through the bottom valve.Another layer of sand, identical to that at the base, was placed on top of the sample and covered with a latex membrane.After the chamber lid was closed, the cone was pushed to its initial position within the sample.The preparation of such a large-scale tailing sample in the chamber is time-consuming and typically requires up to six weeks.

Penetration configuration
Following sample preparation, the total weight of the sample was 3415 kg, with a height of 99.9 cm, which resulted in an overall void ratio of 0.79, as determined by monitoring the load cells beneath the chamber.The specimen was flushed with water for nearly two weeks before consolidation.Subsequently, it was subjected to consolidation with vertical pressure σv of 350 kPa and back pressure σb of 200 kPa, establishing vertical effective stress σ'v of 150 kPa on the sample surface.
In this calibration chamber test, the cone penetration comprised two stages: Standard rate penetration (2 cm/s) in the upper half of the chamber and slow rate penetration (0.02 cm/s) in the lower half of the chamber, with a pore pressure dissipation test conducted in the middle of the chamber.Following the test, the specimen was unloaded and the hydraulic jack, reaction frame, and other equipment were dismantled individually.Sampling was performed to measure the local void ratios within the tailing specimens.

Results
The results obtained from the calibration chamber test are shown in Figure 5, depicting the depth on the vertical axis and tip resistance qt, sleeve friction fs, and excess pore water pressure at the cone shoulder position Δu2 on the horizontal axis.Notably, the total cone resistance qt was adjusted for the unequal area effect using the formula qt = qc + u2 (1−a), where qc represents the measured tip resistance and a denotes the cone area ratio [9].The reference point for the top chamber horizon was set to zero, with the total cone penetration distance spanning 77 cm and ranging from 27 to 104 cm.At a depth of z = 58 cm, the cone was halted to dissipate the pore pressure.
During fast-rate penetration (2 cm/s), the tip resistance appeared to plateau at depths of z = 43-56 cm, ranging from 4.5 to 5.2 MPa, before decreasing slightly to a stable value of 3 MPa during slow-rate penetration (0.02 cm/s).Notably, the tip resistance was higher at a standard rate of 2 cm/s than at a slower rate of 0.02 cm/s.Similarly, sleeve friction followed a comparable pattern, albeit with significantly lower magnitudes, ranging from 30 to 39 kPa for the first penetration and 14.5 to 17.7 kPa for the second penetration, with the plateau emerging later.In Figure 5 (c), the excess pore pressure Δu2 (i.e., u2 minus the back pressure) is depicted, showing a slight increase at the initial penetration, followed by a noticeable negative pore pressure, reaching up to −150 kPa during the first penetration.However, the excess pore pressure Δu2 remained relatively constant and was close to zero until the end of penetration, indicating the likelihood of drained penetration.

Conclusion
A calibration chamber at the University of Toronto was recommended.The test equipment and procedures were detailed, and the results of the chamber test were analyzed, including the tip resistance qt, sleeve friction fs, and excess pore water pressure at the cone shoulder position Δu2.The tip resistance at a standard rate of 2 cm/s exceeded that at a slow rate of 0.02 cm/s.A negative pore pressure was observed in the upper half of the chamber during standard-rate penetration, whereas it approached zero during slow-rate penetration.To comprehensively evaluate partial drainage in tailings, additional calibration chamber tests utilizing variable rates should be conducted.Furthermore, incorporating field CPT measurements for comparison with chamber results is recommended.

Figure 1 .
Figure 1.Components of the calibration chamber at the University of Toronto.

Figure 2 .
Figure 2. The schematic diagram of the control panel.

Figure 3 .
Figure 3. Grain size distribution of tailings in the chamber.

Figure 5 .
Figure 5. Results of calibration chamber test.